Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix: Whisper generate, move text_prompt_ids trim up for max_new_tokens calculation #23724

Merged
merged 1 commit into from
May 24, 2023
Merged

fix: Whisper generate, move text_prompt_ids trim up for max_new_tokens calculation #23724

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 4 additions & 3 deletions src/transformers/models/whisper/modeling_whisper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1633,6 +1633,9 @@ def generate(
)
prompt_ids = prompt_ids.tolist()
decoder_start_token_id, *text_prompt_ids = prompt_ids
# Slicing the text prompt ids in a manner consistent with the OpenAI implementation
# to accomodate context space for the prefix (see https://github.com/openai/whisper/blob/c09a7ae299c4c34c5839a76380ae407e7d785914/whisper/decoding.py#L599)
text_prompt_ids = text_prompt_ids[-self.config.max_length // 2 - 1 :]
# Set the decoder_start_token_id to <|startofprev|>
kwargs.update({"decoder_start_token_id": decoder_start_token_id})

Expand All @@ -1647,9 +1650,7 @@ def generate(
kwargs.pop("forced_decoder_ids", None) or generation_config.forced_decoder_ids
)
forced_decoder_ids = [
# Slicing the text prompt ids in a manner consistent with the OpenAI implementation
# to accomodate context space for the prefix (see https://github.com/openai/whisper/blob/c09a7ae299c4c34c5839a76380ae407e7d785914/whisper/decoding.py#L599)
*text_prompt_ids[-self.config.max_length // 2 - 1 :],
*text_prompt_ids,
generation_config.decoder_start_token_id,
*[token for _rank, token in non_prompt_forced_decoder_ids],
]
Expand Down